Alphaviruses and flaviviruses are the causative agents of severe human and animal illnesses such as encephalitis, polyarthritis, and dengue fever, with millions of cases in humans occurring per year. These enveloped viruses infect cells via a membrane fusion reaction that merges the virus and cell membranes and releases the viral genome into the cytoplasm. This fusion mechanism involves proteolytic cleavage of a companion subunit to activate the fusion protein, fusion protein trimerization via a non-coiled coil mechanism, and insertion of the internal fusion peptide into the target membrane. Fusion of this class of viruses thus is mechanistically quite different from the well characterize influenza virus fusion reaction, in which the fusion protein itself is cleaved and forms a trimeric coiled-coil structure. The goal of this proposal is to determine which the molecular mechanism of membrane fusion in the alphavirus Semliki Forest virus (SFV), a member of this class of viruses and a highly developed system to study membrane fusion. Three key features of the SFV fusion reaction will be addressed: 1. What are the functions of the E1 internal fusion peptide and transmembrane domain in fusion? Photolabeling will be use to define the region of E1 that inserts into the target membrane during fusion, and mutagenesis will test the role of specific residues in this region. The structure of virus containing a fusion-blocking mutation will be determine by cryo electron microscopy. The role of highly conserved glycine residues within the E1 transmembrane domain will be analyzed. 2. How does the E2 companion subunit interact with E1 and regulate its membrane fusion activity? Virus mutants with alterations in E1/E2 dimer stability will be selected and used to map the protein domains that are important in these subunit contacts. The effects of these mutations on fusion activity will be determined. 3. How does the E1 fusion protein oligomerize to form an E1 homotrimer, a critical step in fusion? This aim will characterize the minimal domain of E1 required to trimerization and the protein conformation changed involved in the E1 monomer to trimer transition. Molecular information on the fusion reaction of viruses in this class will make possible the design of specific inhibitors of key early steps in viral infection, and further our understanding of the ubiquitous membrane fusion reactions important to both viruses and cells.